MiniDSP Flex HTx Measurement (FW1.1.17 Update)

[sarieri]

I'm confused by the multi-tone issue...if no filters are applied I would expect no clipping until digital clipping at 0+ volumes. If filters are applied, then the clipping could occur at a negative volume: eg a filter that boosts volume at a frequency by 5db could clip at -5 master volume. Is my understanding correct? In which case I would know at exactly what volume it is "safe" to go to?

You will clip at 0dB for Multitone test. That’s why there’s a crest factor and I use -11dB to test multitone. But even that still seems to clip, that’s why it’s a bug.

 

[sarieri]

A small update. Minidsp is obviously actively working with ESS to look into the issues I mentioned before. They updated me with some progress and hopefully it can be fixed with a firmware update.

 

[danstahl]

You will clip at 0dB for Multitone test. That’s why there’s a crest factor and I use -11dB to test multitone. But even that still seems to clip, that’s why it’s a bug.

This seems like a deal-breaker...the lower the volume the more bits are lost and SINAD/noise is going to take a hit. Unless I'm missing something.

 

[sarieri]

This seems like a deal-breaker...the lower the volume the more bits are lost and SINAD/noise is going to take a hit. Unless I'm missing something.

No. My measurements literally showed that at -10dB or -12dB, the Multitone noise floor(>19bits) is lower than 0dBFS Multitone noise floor(around 16bits). And this has nothing to do with SINAD. You can check my measurements on SINAD vs level graph, it is almost perfect linear.

 

[howard416]

What might help is if you make the x and y axes the same on the before and after versions of the graphs? Initially, I thought the "fixed" performance was worse.

 

[yyyoyo1019]

Hello, I have replaced the original HTX machine with a better input filtering capacitor and coupling capacitor through the bridge matching. Additionally, I have replaced the original OPA1612 operational amplifier with an OPA1656 operational amplifier. Do you think these operations have improved the sound?

 

[sarieri]

Hello, I have replaced the original HTX machine with a better input filtering capacitor and coupling capacitor through the bridge matching. Additionally, I have replaced the original OPA1612 operational amplifier with an OPA1656 operational amplifier. Do you think these operations have improved the sound?

As long as you don’t change any capacitor in the IV circuit, I believe changing them to better caps will have no impact on measurements. Replacing the OPA1612 to OPA1656 may have some effect on measurements. The ESS reference design provides a schematics with OPA1612, I will personally stick with that.

 

[yyyoyo1019]

As long as you don’t change any capacitor in the IV circuit, I believe changing them to better caps will have no impact on measurements. Replacing the OPA1612 to OPA1656 may have some effect on measurements. The ESS reference design provides a schematics with OPA1612, I will personally stick with that.

Opa1656 is indeed slightly inferior to opa1612 in terms of measurement, but it does make people feel more comfortable in terms of sound and music sensation. After all, the HTX index is already high enough. Replacing opa1612 should be an improvement in my subjective listening experience.
If there is indeed a problem after replacement, then I can replace it again.
I did not change any capacitance or specifications for the other capacitors, but selected Panasonic's polymer capacitor through bridge measurement, which has better ESR and D values.

 

[sarieri]

Opa1656 is indeed slightly inferior to opa1612 in terms of measurement, but it does make people feel more comfortable in terms of sound and music sensation. After all, the HTX index is already high enough. Replacing opa1612 should be an improvement in my subjective listening experience.
If there is indeed a problem after replacement, then I can replace it again.
I did not change any capacitance or specifications for the other capacitors, but selected Panasonic's polymer capacitor through bridge measurement, which has better ESR and D values.

Actually, OPA1656 has better linearity comparing to OPA1612 in some use case. I cannot comment on this without a measurement.

 

[yyyoyo1019]

Actually, OPA1656 has better linearity comparing to OPA1612 in some use case. I cannot comment on this without a measurement.

Many of my sound cards use 1656, such as Antelope and Apogee. I have measured the performance of this operational amplifier in the later stage buffer instead of 1612, and it does indeed have better linearity.

 

[sarieri]

Many of my sound cards use 1656, such as Antelope and Apogee. I have measured the performance of this operational amplifier in the later stage buffer instead of 1612, and it does indeed have better linearity.

I have few OPA1656 handy to try. I may try at some point. But right now, i believe the main priority for Htx is looking into the registers of ess9017s. I believe the multitone bug I found is related to that. Also, it might be interesting to learn whether it will be better to seap the 9017s with the 9027pro.

 

[yyyoyo1019]

I have few OPA1656 handy to try. I may try at some point. But right now, i believe the main priority for Htx is looking into the registers of ess9017s. I believe the multitone bug I found is related to that. Also, it might be interesting to learn whether it will be better to seap the 9017s with the 9027pro.

I have sent an email to Minisp asking if it is possible to replace the 9027pro, but they do not recommend me doing so.
As far as I am concerned, I have found through reading the specifications of ESS that the packaging and peripheral circuits of 9017s and 9027pro are almost the same. Perhaps in the future, I will try to replace the DAC chip.

 

[sarieri]

I have sent an email to Minisp asking if it is possible to replace the 9027pro, but they do not recommend me doing so.
As far as I am concerned, I have found through reading the specifications of ESS that the packaging and peripheral circuits of 9017s and 9027pro are almost the same. Perhaps in the future, I will try to replace the DAC chip.

The pins of the two chips are the same but not the reference IV circuit. Also, there might be some parameters that minidsp put there in software mode that won’t work will by just changing the chip.

 

[yyyoyo1019]

ould you try replacing it?

The pins of the two chips are the same but not the reference IV circuit. Also, there might be some parameters that minidsp put there in software mode that won’t work will by just changing the chip.

Would you try replacing it?

 

[sarieri]

ould you try replacing it?

Would you try replacing it?

I don’t think so. I’m more interested in that multitone bug at the moment

 

[yyyoyo1019]

I don’t think so. I’m more interested in that multitone bug at the moment

Do you think that multitone bug is easy to fix?

 

[sarieri]

Do you think that multitone bug is easy to fix?

No. I don’t think so.

 

[yyyoyo1019]

No. I don’t think so.

Does it affect my listening to music?

 

[sarieri]

Does it affect my listening to music?

Changing the chip will definitely change the sound.

 

[yyyoyo1019]

Changing the chip will definitely change the sound.

I am referring to HTX itself without replacing other DAC chips in the case of multi-tone problem, whether it will have an impact on listening to music.